Dynamic Robotic Control Strategies for UAV Stability with Rotatory Payload

Unmanned aerial vehicles (UAVs) offer new possibilities for automating construction tasks, yet handling rotatory payloads such as spools or float pans introduces significant stability challenges due to uneven load distribution and rotational forces. This paper presents a dynamic control framework that enhances UAV stability when managing rotatory payloads. The system integrates real-time sensor data from load cells, IMUs, and gyroscopes with a radial basis function interpolation method to map load distribution and rotational dynamics. A hybrid control strategy combining nonlinear model predictive control (MPC) and reinforcement learning (RL) optimizes motor thrust and payload positioning. MPC provides predictive stability control, while RL enables real-time adaptation to disturbances. Experiments in simulation environment show less than 5% orientation deviation, recovery times under 2.5 s, and a 22% increase in payload capacity. Field tests demonstrate a 26% reduction in task time while maintaining stable performance, validating the system’s potential for dynamic construction operations.